Computers and Electronics in Agriculture 96 (2013) 4056

Contents lists available at SciVerse ScienceDirect

Computers and Electronics in Agriculture

journal homepage: www.elsevier.com/locate/compag

Measuring pressures in a slender cylindrical silo for storing

maize. Filling, static state and discharge with different material ow
rates and comparison with Eurocode 1 part 4
A. Couto , A. Ruiz, L. Herrez, J. Moran, P.J. Aguado
Department of Agricultural Engineering and Sciences, ESTI Agricultural, University of Leon, Av. Portugal 41, 24071 Len, Spain

a r t i c l e

i n f o

Article history:
Received 4 January 2013
Received in revised form 11 April 2013
Accepted 14 April 2013

Keywords:
Test silo
Silo pressures
Maize storage
Discharge ow variation

a b s t r a c t
In this paper, we report the results for pressures in full-size silos obtained from assays performed on a
test station. This basically consisted of a mid-scale cylindrical silo equipped with load cells, with which
it was possible to obtain most of the parameters used in the various theoretical calculation models and in
existing regulations.
Three different types of assay were conducted by varying the mode of silo discharge. In the rst kind of
assay, the silo was completely emptied following a period during which the stored material remained static. In the second type, a small amount of material was unloaded before interrupting the discharge, and in
the third type, discharge was effected in several stages. We also studied the inuence of variations in
pressure on discharge speed.
This study shows that variation in the discharge ow rate is not associated with a further increase in
pressures and that the state of overpressure attained during discharge is essentially due to compaction or
an increase in the specic weight of the material in given areas of the silo. We also observed that the discharge of a small amount of material caused a variation in pressures of the same magnitude as a complete
discharge, and that previous pressure levels were not recovered once a partial discharge had been halted.
The results of this study are compared with the Eurocode 1, part 4; in all cases, the values obtained were
lower than those proposed in the Eurocode.
2013 Elsevier B.V. All rights reserved.

1. Introduction
At present, specialist silo design engineers and companies often
incorporate safety margins into their silo design calculations,
mainly due to the unavailability of reliable real models, as these
are very expensive and the results obtained are only valid for similar silos (Ayuga et al., 2001a; Sadowski and Rotter, 2011). In general, the various elements that make up agricultural silos are often
deliberately oversized, especially as regards thickness, in order to
avoid dangerous situations which may lead to structural collapse
(Juan et al., 2006). Thus, a better understanding of the actions that
affect silos is required, and this can only be achieved through
developing and validating models and theories that facilitate optimal sizing of structural elements in order to improve safety, stability and reliability and to avoid the unnecessary costs derived from
oversizing due to uncertainty. However, very few experimental
installations in the world have full-scale silos (Brown et al.,

Corresponding author. Tel./fax: +34 987 29 10 00x5243.

E-mail addresses: acouy@unileon.es (A. Couto), aruip@unileon.es (A. Ruiz),
luis.herraez@unileon.es (L. Herrez), julia.moran@unileon.es (J. Moran), pedro.aguado@unileon.es (P.J. Aguado).
0168-1699/$ - see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.compag.2013.04.011

2000; Hrtl et al., 2008; Ramirez et al., 2010; Schurich et al.,

2001; Teng and Chan, 2001; Teng et al., 2005; Zhong et al., 2001)
and very few assays have been conducted on them (Askegaard
and Munch-Andersen, 1985; Chen et al., 2005, 2007; Hrtl et al.,
2008; Ooi et al., 1990; Rotter et al., 2002; Teng et al., 2005; Yang
and Hsiau, 2001). Such structures and tests demand a considerable
investment, and although they have yielded extremely important
results, many uncertainties still remain which require further research in order to reliably predict the behaviour of material stored
in this kind of structure.
Several researchers have developed numerical theories and
models to predict the behaviour of material stored inside a silo
(Artoni et al., 2009; Ayuga et al., 2001a; Briassoulis, 2000; Chen
et al., 1998; Ding et al., 2011; Gillie and Holst, 2003; Goodey and
Brown, 2004; Guaita et al., 2003; Juan et al., 2006; Khelil, 2002;
Martinez et al., 2002; Matchett et al., 2008; Song, 2004; Tejchman
and Ummenhofer, 2000; Teng and Chan, 1999; Vidal et al., 2006,
2008; Yang and Hsiau, 2001; Yang et al., 2011); however, most lack
experimental validation. Furthermore, to be reliable, such theoretical models would require the kind of accurate data on certain
parameters of the stored material which are not yet available,
due to insufcient research (Moya et al., 2002; Ramirez et al.,

41

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

2009). Experimental studies have also been conducted, using silos

built to scale (Ahn et al., 2008; Balevicius et al., 2011; Chou et al.,
2002; Coetzee and Els, 2009; Grudzien et al., 2011; Rotter et al.,
1989; Sielamowicz et al., 2005, 2011; Wu et al., 2009); the problem
with these kinds of assay is the subsequent high risk of committing
errors due to scale, since predictions of pressures on the walls of a
full-size silo are based on the results obtained from scaled down silos (Nielsen and Askegaard, 1977).
What is known and widely accepted is that the greatest pressures on silo walls occur during discharge (Askegaard and
Munch-Andersen, 1985; Ayuga et al., 2001b; CEN, 2006; Hrtl
et al., 2008; Ramirez et al., 2010; Zhong et al., 2001), although despite the many studies which have been carried out, determination
of the magnitude and distribution of pressures during discharge
still presents considerable difculties since this is a complex phenomenon in which the pressures depend, among other things, on
the way in which the stored material ows (Chen et al., 2007; Jenike and Johanson, 1969). The ow pattern can be of two types,
mass ow or funnel ow (CEN, 2006), or a mixture of these, and
is essentially governed by the forces acting on the walls. The
parameters which exert most inuence are the angle of the hopper
and the angle of friction between the material and the hopper wall.
In mass ow, all the material moves; this type of ow is typical of
smooth walled hoppers with large outlets and is the most commonly employed in industrial processes because it is the most efcient discharge system and avoids dead zones that can alter the
properties of the material stored (Wjcik et al., 2012).
Furthermore, some authors have observed that the presence of
shear zones inuences the pressures exerted on silo walls (Slominski et al., 2007). These shear zones cause uctuations in the lateral
pressures on the walls and may recur throughout the height of the
silo, causing discontinuities or peaks in the curve of normal pressures on the wall (Wjcik and Tejchman, 2009).
In this article, we present the results of several tests conducted
using an experimental cylindrical silo to determine the thrust
forces exerted by the stored material on the walls. The material
used was maize. Data was collected during lling, in static state
and during discharge, and variations in the ow rate (kg/s) of
material during discharge was also analysed by experimenting
with various discharge ow rates. In addition, we tested the effect
on pressures of performing partial discharges. The results were
compared with values obtained using the European standard for
calculating actions in silos (CEN, 2006).
A mid-scale test station was used to carry out the assays. The
installation basically consists of a mid-scale, cylindrical silo
equipped with load cells to measure pressure, with which it was
possible to obtain most of the parameters governing the behaviour
of material stored in silos and to compare the values obtained with
those obtained using the different calculation models and standards which have been proposed. The silo was designed, built
and validated by the same research team responsible for the present paper, and has been described and validated in detail in two
previous publications (Couto et al., 2012; Ruiz et al., 2012b).
In short, the question of the actions to be taken into account in
the structural design of silos used for the storage of particulate solids has been a subject of enquiry since the end of the 19th century
(Janssen, 1895). Since then, many advances in knowledge have
been made, but many aspects of silo structural design still remain
unresolved (Ayuga, 2008; Dogangun et al., 2009; Nielsen, 2008).

containing the instrumentation necessary for measuring the actions of the stored material. The stored material is transferred between both silos using two screw conveyors powered by electric
motors. Readers who wish to obtain more information about the
test station can consult two previous publications (Couto et al.,
2012; Ruiz et al., 2012b), where the main characteristics are described in greater detail.
2.1.1. Geometry of the experimental silo
The geometry and dimensions of the test silo are shown in
Figs. 1 and 2. The silo is cylindrical, with a central hopper. The silo
body (a vertical cylinder) and the hopper are independent structures which can be dismantled.
The silo walls were constructed from sheets of stainless steel
3 mm thick, whilst the cylinder, hopper and roof reinforcement
rings were 50 mm wide and 10 mm thick. Given the dimensions
and the type of material used for the walls, the silo can thus be
considered a rigid, smooth-walled steel silo. According to the Eurocode classication, it corresponds to a slender silo, since the cylinder height/diameter ratio is equal to 2 (hc/dc = 2).
2.1.2. Measuring horizontal pressures
To measure normal wall pressures, a vertical generatrix was located on the cylinder wall, along which 7 readings were taken at
the different heights indicated in Figs. 13, whilst hopper pressures
were measured immediately below the silo-hopper transition, at
the prolongation of the generatrix mentioned above for measuring
cylinder pressures.
Double bending beam load cells were used to measure normal
wall pressures. To this end, 150  150 mm openings were made
in the wall, into which panels of the same curvature were inserted,

2. Materials and methods

2.1. General description of the installation
The principle components of the test station consist of a silo for
storing the material to be tested and another, experimental silo

Fig. 1. Test silo.

42

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

2.1.3. Measuring vertical forces

In order to measure vertical forces caused by friction of the
stored material against the wall, the cylinder and hopper were connected by mounting the external reinforcing rings of each on the
tension/compression load cells. The rings were very close but did
not come into contact, and thus all vertical pressures were absorbed by the load cells (Figs. 3 and 4).
As in the previous section, more detailed information is provided in the two articles cited above (Couto et al., 2012; Ruiz
et al., 2012b).
2.2. Nomenclature
The nomenclature employed is in accordance with that used in
the Eurocode (CEN, 2006), but this article also includes specic aspects which are not considered in the code. Thus these different
parameters are dened as follows:

Fig. 2. Symbology, dimensions and points for measuring normal cylinder wall
pressures of the test silo.

leaving a small gap around the edges. This clearance was of

approximately 1 mm, sufciently large to prevent friction between
the panels and the wall, and sufciently small to prevent grains
from entering the space and distorting measurement readings.
The panels were attached to the load cells as indicated in Fig. 4.
Hopper wall pressure readings were taken in the same way.
More detailed information related to this section is given in two
previous articles published by the same research team (Couto et al.,
2012; Ruiz et al., 2012b), which describe the load cell model used,
its technical specications and the method used to calibrate the
cells.

A: plan cross-sectional area of vertical walled segment, m2.

F: hopper pressure ratio.
K: lateral pressure ratio.
U: perimeter of the plan cross-section of the vertical walled
segment.
W: weight of stored material, kN.
WE: weight at the end of the stage, kN.
WI: weight after lling, kN.
WII: weight after discharge 1.
WS: weight at the start of the stage, kN.
f: grain ow during discharge, kg/s.
nzSk: mean value of vertical stress per perimeter unit at the silohopper transition, kN/m.
phGA1,t: normal wall pressure at time t on the load cell GA1, kN/
m2.
pvt,t: vertical stress in the stored material at the silo-hopper
transition at time t, kN/m2.
phe,u: horizontal pressure on the cylinder wall during discharge
calculated using the simplied method proposed in Eurocode 1,
part 4, kN/m2.
Pne: normal pressure on hopper wall during discharge calculated using the method proposed in Eurocode 1, part 4, kN/m2.
c: value of the bulk unit weight of particulate solid, kN/m3.
/: Angle of internal friction of particulate solid, .

Fig. 3. Measurement of normal cylinder wall pressures (a) and vertical forces (b).

43

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Test silo
hopper
Slide gate

Gap (5 mm)

Screw
feeder

To storage
silo

Grain
collecting
hopper

Fig. 4. Hopper and hopper screw feeder.

Fig. 5. Pressure-unit weight of the grain.

lm: mean value of coefcient of wall friction between particulate solid and the cylinder wall.
2.3. Description of the tests
Tests were conducted using 100% pure corn (Zea mays). The
properties of this material were previously determined experimentally in the School of Agricultural Engineering (E.S.T.I.A.) laboratories in accordance with the assay methodology proposed in
Eurocode 1, part 4 Annex C (CEN, 2006). The values obtained were:
specic weight (see Fig. 5); angle of repose, 34.22; steel wall friction coefcient, 0.22; and humidity, 14.3 %, and it was these values,
obtained in laboratory tests, that were used to estimate pressures
according to the calculation method proposed in the Eurocode. In
the case of specic weight, the value employed in the calculations
was 7.36 kN/m3, which corresponds to the maximum particle
packing density at a stress level corresponding to the position in
the material stored in the silo where maximum vertical stress after
lling occurs, namely, vertical stress in the stored material at the
silo-hopper transition.
It was also possible to obtain some of the properties of the
stored material, such as mean specic weight, steel wall friction
coefcient and lateral pressure ratio, from the data recorded in
the experimental silo. Thus, the mean specic weight obtained
for the material was 7.354 kN/m3, while the values of lm and K
are shown in Tables 1 and 2 for each of the tests conducted.

Using the granular material described above, 19 tests were conducted, consisting of central loading and discharge of the silo, and
these are described below.
In all tests, silo lling was accomplished using a screw conveyor
ending in a vertical tube which was always centred during lling
and was used to effect free discharge into the silo interior, as
shown in Figs. 1, 5 and 6. This method is commonly employed in
this type of structure when the material stored is grain. A uniform
lling speed (kg/s) was used in all cases, and values for each test
are given in Tables 13.
In all cases, discharge was effected by gravity. A grain collecting
device, separated from the silo by a clearance of 5 mm, was located
beneath the outlet and transported the grain to the storage silo.
This gap was sealed to prevent grain spilling though, but a exible
material was used in order to prevent the vibrations caused by
rotation of the screw conveyor device from being transmitted to
the test silo.
The 19 tests mentioned above were grouped into three types of
assays which differed with regard to how discharge was achieved.
Thus, in Type I assays, following a period in static state, the stored
material in the silo was completely discharged. In Type II assays, a
small amount of stored material, between 1.8% and 2.8% of the total, was discharged before halting the process. The grain was thus
returned to static state before nally emptying the silo completely.
Successive discharges were performed in Type III assays, alternating each discharge with a period of static state. Then, after dis-

44

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Table 1
Summary of results obtained in type I assays (tests 18).
TEST 1

t (s)

f (kg/s)

Normal wall pressures (kPa)

PhGA7

PhGA6

PhGA5

PhGA4

PhGA3

PhGA2

PhGA1

PhGA0

1.10
1.12
1.08
1.21

2.38
2.42
2.34
3.58

2.82
2.85
2.80
3.35

2.93
2.96
2.89
3.26

3.46
3.53
3.45
4.81

3.83
3.92
3.82
4.68

4.05
4.16
4.03
5.78

7.64
7.98
7.65
16.19

Filling
Static

426.38
311.05

3.14
0

Discharge
Weight after lling (kN)

346.60
13.14

3.86

nzSk (kN/m)

pvt,t (kPa)

lm

Gate (/ = 350 mm)

1.30
1.29
1.15
1.71

10.05
10.34
9.78
10.40

0.42
0.43
0.40
0.77

0.27
0.23
0.18
0.30

(Max. values)
(Max. values)
(Min. values)
(Max. values)

TEST 2

Gate (/ = 350 mm)

Filling
Static

418.30
341.85

3.23
0

Discharge
Weight after lling (kN)

347.09
13.23

3.89

Filling
Static

428.38
305.58

3.11
0

Discharge
Weight after lling (kN)

352.17
13.08

3.79

Filling
Static

425.77
323.36

3.12
0

Discharge
Weight after lling (kN)

367.15
13.05

3.62

Filling
Static

420.8
332.6

3.17
0

Discharge
Weight after lling (kN)

517.5
13.10

2.58

Filling
Static

420.8
337.6

3.16
0

Discharge
Weight after lling (kN)

526.6
13.05

2.53

1.36
1.38
1.27
1.29

2.45
2.53
2.45
3.59

2.77
2.82
2.76
3.10

2.84
2.91
2.85
2.90

3.46
3.55
3.46
4.55

3.78
3.89
3.79
4.82

3.77
3.90
3.77
6.66

7.03
7.34
7.04
16.00

1.23
1.37
1.12
1.73

10.13
10.60
9.58
10.52

0.40
0.40
0.36
0.87

0.24
0.24
0.18
0.29

1.05
1.06
1.00
1.01

2.10
2.17
2.10
3.65

2.37
2.39
2.33
2.57

2.84
2.76
2.64
2.61

2.71
2.75
2.68
5.10

2.63
2.66
2.62
4.28

2.68
2.72
2.68
5.98

6.99
7.41
6.98
17.10

0.99
1.04
0.88
1.28

11.00
11.39
10.72
11.29

0.31
0.25
0.24
0.76

0.25
0.23
0.18
0.26

1.21
1.16
1.07
1.06

1.90
1.98
1.90
3.84

2.22
2.26
2.22
2.40

2.24
2.26
2.22
2.17

2.80
2.86
2.81
4.76

2.83
2.86
2.82
4.43

3.01
2.94
2.90
6.48

7.17
7.70
7.17
17.09

0.99
0.99
0.87
1.26

11.15
11.38
10.89
10.33

0.35
0.27
0.25
0.74

0.24
0.22
0.18
0.25

1.07
1.09
1.02
1.02

2.12
2.17
2.12
3.47

2.55
2.58
2.50
2.54

2.58
2.60
2.53
2.54

3.08
3.09
3.02
4.74

2.62
2.55
2.48
4.96

3.35
3.39
3.33
6.83

6.28
6.50
6.27
16.02

1.11
1.17
0.98
1.41

10.57
11.01
10.22
10.17

0.34
0.33
0.30
0.76

0.27
0.24
0.18
0.28

1.15
1.16
1.10
1.11

2.13
2.20
2.13
3.45

2.28
2.32
2.28
2.42

2.69
2.71
2.67
2.68

3.16
3.19
3.13
4.27

2.90
2.82
2.74
4.64

3.12
3.11
3.04
6.57

6.68
6.93
6.67
17.85

1.06
1.11
0.92
1.26

10.75
11.18
10.40
10.97

0.31
0.30
0.27
0.70

0.25
0.23
0.18
0.25

TEST 3

Gate (/ = 120 mm)

TEST 4

(Max. values)
(Max. values)
(Min. values)
(Max. values)
Gate (/ = 60 mm)

416.8
357.1

3.24
0

1887.4
13.23

0.71

417.3
324.6

3.21
0

1938.4
13.14

0.69

1.26
1.29
1.18
1.20

2.34
2.43
2.35
3.78

2.57
2.64
2.57
3.07

2.76
2.83
2.74
2.73

2.88
2.95
2.88
4.14

3.47
3.56
3.48
4.72

3.30
3.41
3.30
6.07

6.19
6.48
6.20
16.13

1.30
1.36
1.18
1.53

9.85
10.31
9.61
9.76

0.34
0.35
0.33
0.72

0.28
0.26
0.18
0.27

1.15
1.17
0.99
1.00

1.94
2.01
1.91
3.68

2.11
2.18
2.07
2.97

2.35
2.49
2.30
2.96

2.65
3.45
2.65
8.00

3.02
4.03
3.02
6.24

2.94
4.07
2.94
8.96

6.02
9.28
6.01
19.06

1.03
1.10
0.90
1.25

10.92
11.33
10.55
10.52

0.28
0.39
0.26
1.00

0.25
0.24
0.18
0.20

TEST 8

Discharge
Weight after lling (kN)

(Max. values)
(Max. values)
(Min. values)
(Max. values)
Gate (/ = 90 mm)

TEST 7

Filling
Static

(Max. values)
(Max. values)
(Min. values)
(Max. values)
Gate (/ = 90 mm)

TEST 6

Discharge
Weight after lling (kN)

(Max. values)
(Max. values)
(Min. values)
(Max. values)
Gate (/ = 120 mm)

TEST 5

Filling
Static

(Max. values)
(Max. values)
(Min. values)
(Max. values)

(Max. values)
(Max. values)
(Min. values)
(Max. values)
Gate (/ = 60 mm)

charging approximately 12% of the material, the silo was relled

and the process was repeated before a nal rell and complete single discharge.
The rationale for performing these three types of assay was to
mimic the common use made in practice of these structures, since
although silos are sometimes completely emptied in a single step
after lling, discharge is also frequently halted after partial emptying of the stored material, or the silo may even be relled before it
has been completely emptied of the material it contains. Simulation of the actual use made of these structures has enabled us to
obtain novel ndings which we describe in the following
paragraphs.
2.3.1. Type I assays
In a rst set of assays (Type I, tests 18), the silo was loaded
with granular material and then, after a pause of approximately

(Max. values)
(Max. values)
(Min. values)
(Max. values)

5 min, the material was discharged varying the grain ow rate

(kg/s) between tests. To unload the material, the motors driving
the screw conveyors were switched on and the outlet slide gate
was opened. The hopper feeder (Figs. 4 and 6) remained free of
material in all the tests.
As shown in Fig. 2, the diameter of the outlet was 35 mm, and
this opening corresponded to the maximum ow of granular material tested. To reduce the ow of grain through the silo outlet, different round plates with a circular central orice of varying
diameters were placed over the opening and closing gate (Fig. 4).
Thus tests were carried out with the outlet fully open (gate /
= 350 mm), and with openings of / = 120 mm, / = 90 mm and /
= 60 mm.
The duration of each stage (loading - static - discharge), diameter of the outlet opening and discharge ow rate for each of these

45

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Table 2
Summary of results obtained in type II assays (tests 918).
TEST 9

t (s)

f (kg/s) Normal wall pressures (kPa)

Filling
Static 1

441.8
304.5

3.03
0

Discharge 1
Static 2

3.0
307.7

12.71
0

nzSk (kN/m) pvt,t (kPa) K

lm

Gate (/ = 350 mm)

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

PhGA7 PhGA6 PhGA5 PhGA4 PhGA3 PhGA2 PhGA1 PhGA0

Discharge 2
348.6
WI: Weight after lling (kN)
13.12
WII: Weight after Discharge I (kN) 12.74
WIWII (kN)
0.37

3.73

1.07
1.09
1.02
1.27
1.19
1.00
1.05

2.05
2.12
2.06
3.01
3.15
2.79
3.90

2.19
2.23
2.19
3.08
2.99
2.63
2.80

2.19
2.22
2.17
3.54
3.45
2.55
2.40

2.47
2.53
2.48
4.67
4.58
3.94
5.80

2.53
2.57
2.52
4.97
4.85
4.02
3.90

2.66
2.72
2.66
5.90
5.78
4.79
4.91

7.62
8.05
7.63
17.17
18.47
16.41
18.32

0.95
1.02
0.83
1.23
1.35
0.96
1.44

11.19
11.59
10.86
10.86
10.57
9.04
8.83

0.32
0.25
0.23
0.61
0.59
0.49
0.64

0.24
0.24
0.20
0.30
0.23
0.15
0.27

1.29
1.27
1.15
1.15
1.10
0.74
0.91

2.67
2.62
2.52
3.27
3.15
2.63
4.01

2.81
2.74
2.61
3.40
3.35
2.44
3.07

2.77
2.71
2.57
3.61
3.55
2.64
2.63

2.36
2.39
2.34
4.86
4.82
3.67
6.17

2.96
2.82
2.76
4.80
4.71
3.40
3.31

2.85
2.88
2.82
6.25
6.21
4.16
4.68

7.86
8.25
7.85
17.48
17.37
16.36
18.38

0.99
1.01
0.83
1.15
1.50
0.73
1.57

11.14
11.61
10.89
11.06
11.54
8.51
9.64

0.39
0.26
0.24
0.64
0.67
0.40
0.00

0.22
0.21
0.18
0.23
0.22
0.13
0.32

0.97
0.98
0.92
0.93
0.75
0.70
0.75

2.28
2.32
2.28
3.00
3.05
2.90
3.58

2.17
2.19
2.16
2.18
1.92
1.88
2.55

2.38
2.41
2.35
2.37
1.66
1.61
2.67

2.93
2.97
2.93
5.18
4.94
4.64
5.69

2.67
2.70
2.67
4.53
4.32
3.93
4.06

2.95
2.96
2.92
5.75
5.22
4.88
5.80

6.77
7.00
6.77
16.92
16.91
16.35
16.53

1.00
1.05
0.89
1.41
1.28
0.97
1.41

10.86
11.28
10.65
11.09
10.62
9.37
9.70

0.33
0.28
0.26
0.64
0.53
0.46
0.75

0.24
0.23
0.20
0.28
0.23
0.17
0.25

1.12
1.13
1.05
1.06
0.83
0.77
0.88

2.04
2.10
2.05
2.86
2.88
2.70
3.70

2.66
2.63
2.54
2.54
2.20
2.17
2.50

3.29
3.25
3.13
3.13
1.89
1.83
2.03

3.27
3.25
3.16
4.80
4.77
4.53
5.51

3.06
3.02
2.93
4.52
3.86
3.59
3.89

3.58
3.45
3.27
5.65
5.29
4.90
5.69

7.32
7.70
7.31
17.59
17.58
17.02
17.40

1.04
1.02
0.87
1.31
1.29
0.93
1.28

11.00
11.35
10.74
10.74
10.77
9.31
10.15

0.35
0.32
0.29
0.61
0.54
0.46
0.72

0.22
0.20
0.17
0.24
0.23
0.16
0.24

1.07
1.08
1.01
1.02
0.82
0.75
0.84

2.22
2.29
2.23
3.19
3.16
2.98
3.68

2.34
2.36
2.33
2.37
2.19
2.16
2.69

2.68
2.69
2.64
2.64
1.70
1.60
2.02

3.33
3.27
3.18
4.31
4.27
4.07
4.94

2.90
2.82
2.75
4.46
4.36
4.05
4.30

3.15
3.14
3.08
6.11
5.88
5.45
6.39

6.89
7.13
6.90
17.25
17.24
16.82
17.22

1.05
1.06
0.90
1.22
1.26
0.91
1.24

10.90
11.23
10.61
11.06
10.87
9.46
10.51

0.32
0.30
0.27
0.63
0.60
0.50
0.69

0.29
0.22
0.19
0.24
0.21
0.16
0.23

1.08
1.08
1.01
1.02
0.78
0.73
0.80

2.04
2.12
2.04
2.97
2.98
2.81
3.52

2.35
2.37
2.33
2.35
2.09
2.07
2.52

2.81
2.75
2.68
2.70
1.78
1.65
2.03

2.86
2.90
2.85
3.56
3.55
3.36
4.52

2.76
2.78
2.75
4.47
4.04
3.73
4.10

3.31
3.28
3.21
6.05
5.72
5.31
6.19

6.80
7.05
6.81
18.50
18.48
18.03
18.22

1.01
1.05
0.89
1.19
1.25
0.89
1.21

10.95
11.28
10.64
10.94
10.93
9.50
9.80

0.35
0.31
0.28
0.62
0.58
0.49
0.69

0.23
0.22
0.19
0.24
0.23
0.16
0.23

TEST 10

Gate (/ = 350 mm)

Filling
Static 1

431.8
337.2

Discharge 1
Static 2

2.0
17.49
405314.7 0

Discharge 2
350.1
WI: Weight after lling (kN)
13.12
WII: Weight after Discharge I (kN) 12.77
WI-WII (kN)
0.34

3.10
0

3.72

TEST 11

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 120 mm)

Filling
Static 1

427.8
306.2

3.11
0

Discharge 1
Static 2

5.5
303.2

5.11
0

Discharge 2
348.6
WI: Weight after lling (kN)
13.05
WII: Weight after Discharge I (kN) 12.78
WIWII (kN)
0.28

3.74

TEST 12

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 120 mm)

Filling
Static 1

426.9
309.1

3.11
0

Discharge 1
Static 2

5.5
293.6

5.11
0

Discharge 2
355.2
WI: Weight after lling (kN)
13.04
WII: Weight after Discharge I (kN) 12.77
WIWII (kN)
0.28

3.66

TEST 13

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 90 mm)

Filling
Static 1

421.8
379.2

3.15
0

Discharge 1
Static 2

10.5
314.6

2.60
0

Discharge 2
510.4
WI: Weight after lling (kN)
13.04
WII: Weight after Discharge I (kN) 12.78
WIWII (kN)
0.27

2.55

TEST 14

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 90 mm)

Filling
Static 1

421.8
320.6

3.15
0

Discharge 1
Static 2

10.0
314.1

2.71
0

Discharge 2
512.9
WI: Weight after lling (kN)
13.03
WII: Weight after Discharge I (kN) 12.77
WIWII (kN)
0.27

2.54

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

(continued on next page)

46

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Table 2 (continued)
TEST 9

t (s)

f (kg/s) Normal wall pressures (kPa)

nzSk (kN/m) pvt,t (kPa) K

lm

1.00
1.03
0.88
1.18
1.30
0.79
1.27

0.24
0.22
0.19
0.23
0.23
0.15
0.23

Gate (/ = 350 mm)

PhGA7 PhGA6 PhGA5 PhGA4 PhGA3 PhGA2 PhGA1 PhGA0

TEST 15

Gate (/ = 90 mm)

Filling
Static 1

432.3
341.2

3.07
0

Discharge 1
Static 2

10.5
54168.0

2.57
0

Discharge 2
500.4
WI: Weight after lling (kN)
13.04
WII: Weight after Discharge I (kN) 12.77
WIWII (kN)
0.27

2.60

1.46
1.39
1.26
1.26
0.94
0.78
0.96

2.49
2.44
2.39
3.59
3.57
3.06
3.96

2.18
2.20
2.17
2.17
2.05
1.92
2.68

2.36
2.40
2.36
2.37
1.56
1.34
2.06

3.06
3.11
3.05
3.77
3.77
3.31
4.99

2.54
2.59
2.54
4.70
4.58
3.77
4.61

3.05
3.10
3.05
5.79
5.66
4.54
5.92

7.08
7.25
7.07
18.13
17.95
16.87
17.00

10.95
11.33
10.71
10.87
11.32
9.30
9.73

0.35
0.29
0.27
0.58
0.59
0.40
0.87

TEST 16

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 60 mm)

Filling
Static 1

411.8
332.1

3.24
0

Discharge 1
Static 2

34.6
373.3

0.77
0

Discharge 2
1813.9
WI: Weight after lling (kN)
13.11
WII: Weight after Discharge I (kN) 12.85
WIWII (kN)
0.26

0.72

1.13
1.16
1.04
1.06
0.83
0.71
0.90

2.06
2.14
2.06
3.15
3.21
2.91
3.71

2.36
2.42
2.36
2.99
2.74
2.41
3.03

2.44
2.50
2.44
2.93
2.43
2.04
2.43

2.79
2.88
2.79
5.25
5.44
4.82
6.47

3.16
3.24
3.16
6.32
6.27
5.60
5.59

3.30
3.36
3.29
5.12
4.66
4.42
5.02

6.03
6.22
6.00
17.02
17.01
16.18
16.99

1.10
1.18
0.96
1.27
1.31
0.97
1.25

10.58
11.08
10.21
10.46
10.69
9.35
9.73

0.97
0.32
0.30
0.53
0.49
0.41
0.57

0.25
0.24
0.20
0.23
0.20
0.15
0.21

1.17
1.19
1.10
1.12
1.19
0.81
0.87

1.91
2.00
1.92
3.08
3.08
2.88
3.73

2.26
2.30
2.26
2.86
2.86
2.56
2.73

2.22
2.29
2.21
2.81
2.81
2.42
2.44

2.55
2.63
2.55
6.81
6.81
6.24
7.02

2.67
2.74
2.65
6.27
6.27
5.68
9.84

2.91
2.96
2.90
4.72
4.72
4.33
4.85

5.79
6.00
5.77
16.79
16.79
16.31
17.03

1.06
1.13
0.92
1.24
1.24
0.95
1.24

10.84
11.32
10.48
11.13
11.32
9.58
10.10

0.98
0.28
0.26
0.48
0.48
0.40
0.60

0.27
0.26
0.21
0.24
0.27
0.14
0.17

TEST 17

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 60 mm)

Filling
Static 1

426.4
312.5
312.5
Discharge 1
34.6
Static 2
325.2
325.2
Discharge 2
1831.4
WI: Weight after lling (kN)
13.19
WII: Weight after Discharge I (kN) 12.93
WIWII (kN)
0.26

3.15
0
0.76
0
0.72

TEST 18

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)

Gate (/ = 60 mm)

Filling
Static 1

418.3
344.1

3.19
0

Discharge 1

32.6

0.76

1.10
1.12
1.05
1.04

1.90
1.97
1.91
3.01

2.11
2.15
2.11
2.86

2.18
2.21
2.16
2.86

diameters in the different tests are given in Table 1. Fig. 6 shows a

diagram of the tests.

2.3.2. Type II assays

As with the previous set of assays described, the Type II assays
(tests 918) were performed by varying the grain ow rate (kg/s)
in the same way as explained for Type I assays; however, in these
tests, discharge was performed in two stages. In the rst stage, the
grain outlet gate was opened with the discharge device motors
switched off, so that unloading was interrupted once the hopper
discharge feeder had become lled to capacity, and the granular
material returned to static state. The dwell time in this stage varied
between the different tests. In the second stage, discharge was resumed by switching on the motors once the hopper discharge feeder was full. Fig. 7 shows a diagram of this type of test. Dwell times
in each of the stages, the diameter of the outlet opening, discharge
ow rate and quantity of material discharged in each stage for the
various tests can be found in Table 2.

2.3.3. Type III assay

Lastly, we performed the Type III assay (test 19), consisting of
loading the silo, performing a partial discharge of approximately

2.47
2.54
2.45
6.97

2.90
2.95
2.89
6.67

2.89
2.93
2.87
6.49

5.49
5.72
5.49
15.60

1.01
1.03
0.89
1.23

11.00
11.34
10.77
10.84

0.97
0.27
0.26
0.77

0.26
0.24
0.20
0.23

(Max. values)
(Max. values)
(Min. values)
(Max. values)

1.5 kN and then reloading to the initial level, repeating the process
2 times.
In this test, the outlet opening was 120 mm, and the discharge
device motors were running when this was opened; consequently,
the hopper discharge feeder did not become blocked at any time.
The two partial discharges were performed in ve stages of
0.3 kN each. Once 0.3 kN had been discharged, unloading was
interrupted by closing the outlet gate and the material remained
in static state for 5 min before repeating the process another four
times until all 1.5 kN had been discharged. The silo was then relled to the initial level, and further partial discharges were performed as described above, but this time in stages of 1.5 kN.
Lastly, the silo was once again relled, and complete discharge
was performed in a single step. The data obtained from this assay
are shown in Table 3.

3. Results and discussion

In this section, we report and discuss the results of the three
types of assays performed following the procedures outlined in
the previous section. The shape of the curves obtained for each
set of Type I and Type II assays was similar, and only the dwell time
in each of the static-discharge stages and the maximum and min-

47

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Open
gate

Closed
gate

Closed
gate

Empty

Empty

Full
2.- STATIC

1.- FILLING

Discharge
To storage silo
3.- DISCHARGE

Fig. 6. Diagram of Type I assays.

Table 3
Summary of results obtained in the type III assay (test 19).
TEST 19

t (s)

tT (s)

Ws (kN) WE (kN) f (kg/s) Normal wall pressures (kPa)

nzSk (kN/m) pvt,t (kPa) K

Gate (/ = 120 mm)

PhGA7 PhGA6 PhGA5 PhGA4 PhGA3 PhGA2 PhGA1 PhGA0

Filling 1
417.8 417.8
After lling 1 310.6 728.3

0.000
12.699

12.699
12.699

3.10
0

Discharge 1
Static 1

7.5
735.9
12.699
297.1 1032.9 12.375

12.375
12.375

4.39
0

Discharge 2
Static 2

7.0
1039.9 12.375
302.1 1342.0 12.052

12.052
12.052

4.71
0

Discharge 3
Static 3

6.0
1348.0 12.052
302.6 1650.6 11.775

11.775
11.775

4.69
0

Discharge 4
Static 4

5.5
1656.1 11.775
303.6 1959.7 11.468

11.468
11.468

5.68
0

Discharge 5
Static 5

7.0
1966.7 11.468
292.0 2258.7 11.166

11.166
11.166

4.38
0

Filling 2
52.1
2310.8 11.166
After lling 2 307.1 2617.9 12.669

12.669
12.669

2.94
0

Discharge 6
Static 6

6.5
2624.4 12.669
302.5 2927.0 12.351

12.351
12.351

4.98
0

Discharge 7
Static 7

5.5
2932.5 12.351
303.1 3235.6 12.071

12.071
12.071

5.19
0

Discharge 8
Static 8

6.5
3242.1 12.071
302.5 3544.6 11.780

11.780
11.780

4.56
0

Discharge 9
Static 9

6.0
3550.6 11.780
304.7 3855.3 11.474

11.474
11.474

5.19
0

Discharge 10
Static 10

6.0
3861.4 11.474
301.0 4162.4 11.182

11.182
11.182

4.96
0

Filling 3
53.6
4215.9 11.182
After lling 3 306.6 4522.6 12.680

12.680
12.680

2.85
0

Discharge 11

0.000

3.79

341.0 4863.6 12.680

1.07
1.06
0.92
0.88
0.60
0.56
0.60
0.52
0.48
0.54
0.47
0.44
0.53
0.49
0.44
0.48
0.29
0.26

1.70
1.76
1.70
2.57
2.55
2.41
2.95
2.93
2.74
3.03
3.02
2.80
3.15
3.20
2.93
3.20
3.13
2.94

2.25
2.27
2.17
2.28
2.16
2.13
2.20
2.02
1.95
2.06
1.99
1.92
1.94
1.87
1.80
2.01
2.01
1.95

2.18
2.21
2.08
1.95
1.62
1.59
2.64
1.59
1.57
1.58
1.54
1.47
2.59
1.84
1.80
1.81
1.61
1.58

2.62
2.65
2.56
4.67
4.47
4.22
5.53
5.45
5.14
5.71
5.29
5.06
5.39
5.07
4.88
5.37
4.99
4.82

2.98
2.84
2.69
3.71
3.46
3.16
3.65
2.93
2.76
3.26
2.91
2.72
3.32
2.75
2.60
3.14
2.91
2.75

2.71
2.74
2.54
5.04
4.89
4.49
5.05
4.76
4.43
5.13
4.82
4.53
4.76
4.57
4.29
4.60
4.55
4.28

7.37
9.95
7.37
14.80
14.82
14.46
14.69
14.46
14.05
14.34
13.92
13.52
13.99
14.04
13.59
14.02
13.30
12.87

0.96
1.11
0.85
1.37
1.21
0.90
1.41
1.26
0.90
1.33
1.24
0.90
1.31
1.22
0.87
1.24
1.15
0.83

10.56
11.00
9.96
9.81
10.39
9.15
9.93
9.98
8.53
9.94
9.60
8.23
9.36
9.33
7.93
9.11
9.11
7.83

0.47
0.26
0.25
0.56
0.51
0.43
0.63
0.54
0.44
0.65
0.57
0.47
0.62
0.56
0.46
0.58
0.56
0.47

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)

0.92
0.93
0.86
0.87
0.73
0.67
0.74
0.63
0.56
0.63
0.60
0.56
0.57
0.46
0.42
0.46
0.35
0.32

3.13
2.96
2.92
3.71
3.74
3.53
3.82
3.83
3.60
4.05
4.04
3.77
4.06
3.94
3.71
3.76
3.57
3.36

2.75
2.71
2.65
2.80
2.69
2.60
2.93
2.94
2.84
2.85
2.68
2.62
2.89
2.85
2.74
3.18
2.79
2.71

1.85
1.89
1.85
3.25
2.07
2.04
2.57
1.94
1.91
1.94
1.74
1.69
2.99
2.26
2.22
3.06
1.99
1.95

4.90
4.91
4.84
5.28
5.23
5.07
5.42
5.35
5.16
5.55
5.07
4.93
5.33
5.23
5.06
5.18
5.16
4.95

2.88
2.90
2.87
3.91
3.21
3.05
4.16
3.21
3.02
3.75
3.22
3.05
3.64
3.25
3.05
3.99
2.91
2.73

4.32
4.35
4.31
4.98
4.95
4.64
4.97
4.73
4.42
4.97
5.31
4.94
5.11
5.02
4.73
4.88
4.67
4.35

13.47
13.44
13.41
14.35
13.94
13.45
13.83
13.53
13.00
13.50
13.79
13.27
13.78
12.88
12.43
12.96
12.96
12.55

1.08
1.11
0.96
1.50
1.35
1.00
1.42
1.34
0.96
1.35
1.29
0.92
1.35
1.27
0.91
1.28
1.24
0.88

10.24
10.51
9.91
9.91
9.93
8.52
9.34
9.73
8.20
9.41
9.53
8.05
9.43
9.19
7.73
8.82
8.65
7.49

0.49
0.44
0.41
0.60
0.56
0.47
0.61
0.56
0.45
0.60
0.63
0.52
0.67
0.63
0.52
0.64
0.60
0.43

(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)
(Max. values)
(Max. values)
(Min. values)

1.15
1.11
1.02
1.03

3.56
3.55
3.48
4.48

2.86
2.91
2.86
3.45

2.14
2.20
2.15
2.71

5.06
5.05
4.99
5.95

2.89
2.93
2.88
4.24

4.47
4.41
4.33
5.76

13.35
13.51
13.25
14.43

1.24
1.15
0.98
1.47

9.93
10.44
9.76
9.75

0.60
0.45
0.42
0.73

(Max. values)
(Max. values)
(Min. values)
(Max. values)

48

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Filling

Empty

Empty
1.- FILLING

Open
gate

Open
gate

Closed
gate

Closed
gate

2.- STATIC 1

Open
gate

Full
3.- DISCHARGE 1

Full
4.- STATIC 2

Discharge
To storage silo
5.- DISCHARGE 2

Fig. 7. Diagram of Type II assays.

Fig. 8. Normal silo wall pressures (phGA,t) and vertical stress in the stored material at the transition (pvt,t), at each time t. Type I assays, test 6.

imum values attained varied. Thus, in order to avoid an excessive

number of gures, only the curves corresponding to one test of
each type are shown.
Tables 13 provide a summary of the data obtained for each
test, showing the duration of the various stages of each test, the
ow rate of the discharged material, the weight of the material
stored in each stage and a summary of the results obtained for each
of the stages.
In all the tests it was observed that during discharge, all of the
stored material was moving and thus the results correspond to a
mass ow pattern.

Fig. 8 shows the shape of the curves obtained for Type I assays
(tests 18), and gives normal silo wall pressures and vertical stress
in the stored material at the silo-hopper transition at each instant.
Also included is a curve showing the evolution of the weight of the
stored material throughout the test, and Table 1 provides a summary of the conditions and results obtained in the Type I assays.
As expected, normal wall pressures underwent considerable
variation at the beginning of discharge, reaching highest values
on the hopper wall at the level of the silo-hopper transition, which
is consistent with European standards (CEN, 2006) and with international tests on real silos (Askegaard and Munch-Andersen, 1985;

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Hrtl et al., 2008; Ramirez et al., 2010; Zhong et al., 2001). As can
be deduced from Table 1, there was an increase of up to 157.6% at
the hopper wall and of up to 120.3% in the lower section of the cylinder. It can also be seen that the increase in pressure recorded by
the cells located in the lower section of the cylinder (cells GA0 to
GA3) was always positive, whereas at some points in the upper
section of the cylinder, the pressure remained almost constant or
even decreased with respect to static state. This suggests that as
the grain moved downwards, the density of the material with respect to initial values increased in some areas of the silo and decreased in others. The areas where pressure decreased
constituted discontinuities or rupture zones caused because the
material in the lower layers was descending due to discharge while
the material in the upper layers descended more slowly due to friction with the wall. In a silo of innite height, these rupture zones
mentioned above would recur at regular intervals along the entire
height of the silo. In other words, zones of increased pressure
would alternate with zones of decreased pressure throughout the
entire height of the silo. This is consistent with the results obtained
by Wojcik (Wjcik and Tejchman, 2009), who performed tests on a
silo lled with sand and detected pockets in the initially dense
sand during discharge, which alternated in height along the cylinder wall and took the shape of an arrow pointing to the upper section of the silo.
Fig. 8 also shows that in static state, vertical pressure on the
stored material at the level of the silo-hopper transition increased
over time, as did the normal hopper wall pressure. The opposite occurred in the case of normal cylinder wall pressures, as these decreased over time. In other words, in static state, pressures on
the cylinder walls decreased, increasing the weight resting on the
hopper, and this was due to resettling of the grain in the silo, a phenomenon which has already been discussed in more detail in a previous publication (Ruiz et al., 2012b).
Focusing on ow during discharge, it can be seen that the ow
rate was practically the same as that observed in the tests conducted with 350 and 120 mm diameter outlet openings. This was
because this ow rate coincided with the maximum that the discharge device was capable of unloading. In tests conducted using
an outlet diameter of 350 mm, the ow was much greater during
the initial moments because free discharge occurred, producing a
higher ow than that which the conveyor could transport. Subsequently, the hopper discharge feeder became full and discharge
then proceeded at a constant ow at the rate that the discharge de-

49

vice was capable of evacuating. In tests conducted using an outlet

diameter of 120 mm, the ow of grain into the hopper discharge
feeder was slightly lower than the maximum that the screw conveyer was capable of evacuating, and thus the feeder did not become clogged. In tests conducted using outlet diameters of 90
and 60 mm, the ow rate was again lower than that which the
screw conveyer was capable of evacuating, and thus the feeder
once again remained unclogged.
Fig. 9 shows the normal maximum wall pressures registered
during discharge at the level of the silo-hopper transition according to discharge ow rate in Type I assays. It can be seen that the
pressures on the hopper wall (GA0) decreased linearly as discharge
speed increased (r2 = 0.179). Similarly, the same occurred in the
case of maximum pressures reached on the cylinder wall (GA1),
where pressures were practically constant and independent of
the discharge rate (r2 = 0.326). Thus, according to our results and
contrary to what might be expected, a rapid discharge does not imply higher overpressures on the silo wall.
Fig. 10 shows the evolution of pressures on the wall during discharge for each of the four outlet openings tested, in comparison
with the previous static state and with the pressures obtained by
the calculation method proposed in the Eurocode (CEN, 2006).
Since data on the evolution of pressure was collected approximately half-way through discharge of the stored material, different
time intervals are shown, as discharge was slower in the tests conducted using outlet openings of / = 90 mm and / = 60 mm. It can
be seen in Fig. 10 that pressures on the wall were subject to variation at the start of discharge. This variation was more marked
and always positive in the lower section of the silo at the level of
the silo-hopper transition. It can also be seen that these variations
in pressure did not occur progressively along the height of the silo,
but rather formed alternating peaks in pressure due to the rupture
zones or discontinuities discussed above, a phenomenon which has
also been observed by other authors (Sugita, 1972; Wjcik and
Tejchman, 2009). Comparing the results obtained in the tests with
the values estimated using the calculation method proposed in the
Eurocode (CEN, 2006), it can be seen that the pressures obtained in
the tests were always lower.
Fig. 11 shows the evolution over time of the other parameters
involved in the calculation of silos according to the European standard (CEN, 2006). The value of K is given for the silo loading stage,
from the moment when hopper loading had ended and lling of
the cylinder had begun, and throughout discharge until the silo

Fig. 9. Regression lines obtained from the maximum normal pressures on the cylinder wall and the hopper at the silo-hopper transition recorded in the tests, for each
discharge ow rate. Type I assays.

50

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Fig. 10. Evolution of pressures during discharge for the outlet openings tested. Comparison with Eurocode 1, part 4. Type I assays.

Fig. 11. Mean value of resultant vertical stress per unit perimeter at the transition (nzSk,t), lateral pressure ratio at the transition (K), mean value of the wall friction coefcient
between the stored material and the cylinder wall (lm). Comparison between the values obtained in test 6 and those obtained using Eurocode 1, part 4. Type I assays.

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

had been completely unloaded, intervals at which this parameter is

relevant. As regards the value of l, this is given for loading from
the moment when the level of grain rose above GA7, i.e., from
the moment that the forces registered by this cell related to pressures on the wall, since the grain acts on the entire surface of the
wall connected to this cell, until the grain had dropped again to
the same level during discharge. As in the case of wall pressures,
it can be seen in the gure that the value of nzsk in static state
was not constant but decreased. This is again explained by the phenomenon discussed above of grain resettlement, whereby pressures on the cylinder walls decreased in static state, increasing
the weight resting on the hopper. This phenomenon has previously
been discussed by Sugita (Sugita, 1972), who reported that for 5
or 6 h after loading the silo, pressures in the upper sections decreased and increased on the sloping wall of the hopper.
Fig. 11 also shows the increase in the value of K during discharge, which is logical since the vertical pressure on the stored
material dropped at the beginning of discharge while the normal
cylinder wall pressure at the silo hopper transition increased. A
comparison is also given with the results obtained using the Eurocode (CEN, 2006), showing that the value of nZSK was below that given in the European standard whereas the proposed value of K was
exceeded during discharge.
If we compare the maximum variation in these parameters obtained during discharge with the maximum values obtained in static state, it can be seen that these parameters presented an increase
during the rst moments of discharge, which was more pronounced in the case of the mean value of resultant vertical stress
per unit perimeter at the transition (nzSk,t) and of the lateral pressure ratio at the transition (K), reaching maximum values of
31.5% (test 1) and 159.6% (test 8), respectively.
Fig. 12 shows the shape of the curves obtained in the Type II assays, in which, as explained in Section 2.3.2, discharge was performed in two stages. The results obtained for each stage are
given in Table 2. In the loading stage and the rst stage in static
state, the results obtained in Type I and II assays were comparable

51

and similar: consequently, in order to avoid repetition, these will

not be discussed again here.
As can be seen in Table 2, during the initial discharge, between
0.34 and 0.37 kN of maize was discharged (about 3% of the stored
material) in tests conducted using an outlet opening of /
= 350 mm, whereas in the tests using outlet openings of /
= 120 mm, 90 mm and 60 mm, around 0.240.28 kN was discharged (about 2% of total). The explanation for this difference is
that when the outlet was fully open (/ = 350 mm), the hopper discharge device was completely lled, whereas when all other outlet
opening diameters were used, discharge was effected through a
small hole which the material eventually plugged, at which point
the ow of grain was halted.
The curves shown in Fig. 12 also indicate that once the initial
discharge was halted, normal wall pressures during the second static stage did not return to the previous static state values. See, for
example, the results of tests 10, 15 or 18, in which the second static
state was maintained in some cases for over 112 h without returning to initial pressure values.
This phenomenon should be taken into account when developing new numerical theories and methods for calculating pressures
during silo discharge because it demonstrates that the movement
of material inside the silo due to a small discharge induces changes
in the solid unit weight of the material, which increases or decreases in certain zones along the height of the silo. Thus, the pressure in some zones increases with respect to the previous static
state whilst in others it remains almost constant or even decreases.
This nding supports the theory that the main pressure variations
that occur inside the silo during discharge are essentially due to
changes in unit weight rather than to an increase in volume caused
by lateral movement of the particles during discharge, a phenomenon known as dilatancy, because if the latter were the case, initial
pressure values would be recovered when discharge was halted. A
similar phenomenon has been described by Pieper and Wenzel
(Pieper and Wenzel, 1964), who reported that When discharge
was stopped, the pressure recorded in that moment was main-

Fig. 12. Normal silo wall pressures (phGA,t) and vertical stress in the stored material at the transition (pvt,t), at each time t. Type II assay, test 12.

52

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Fig. 13. Regression lines obtained from the maximum normal pressures on the cylinder wall and the hopper wall at the silo-hopper transition registered during the tests
(discharge 1 and discharge 2), for each discharge ow rate. Type II assays.

tained for about 15 h, although unlike us, they did not state that
initial pressures were not recovered. Their experiments were carried out using cylindrical metal silos measuring 23.6 in. (60 cm)
in diameter and 121 in. (3020 5 cm) high, employing grain and sand
as the stored material.
It can also be seen in Fig. 12 that the resettling of grain inside
the silo in static state, previously discussed for Type I assays, continued to occur in the second static state, which explains the
abrupt steps present in the curves. This gure also includes the
curve for the weight of the material stored, for the different stages.
As can be seen, such steps do not occur in this curve and this, together with other reasons which we have reported previously elsewhere (Ruiz et al., 2012a), indicates that the resettling was not due
to interference in electronic readings.
If we analyse the maximum variation in normal wall pressure
during each of the discharges with respect to the previous static
state, as in the previous tests, the normal wall pressure presented
a substantial variation at the beginning of the rst discharge,
reaching values and pressure distributions inside the silo similar
to those described for Type I assays. However, the variation in pressure following the second discharge was much smaller, with zones
in which it even decreased, and, as will be seen in the Type III assay, initial pressures were recovered on returning to static state.
Consequently, we attribute the variation in pressure during the
second discharge to dilatancy.
In the Type II assays, in contrast to what happened in the Type I
assays, in which the increase in pressure recorded by the cells in
the lower section of the cylinder (cells GA0 to GA3) was always positive whereas in the upper section of the cylinder the pressure remained almost constant or even decreased at some points, in these
assays, and more specically in tests 17 and 19, the GA3 cells began to register pressures below those recorded for static state.
Fig. 13 gives the maximum normal wall pressures in the lower
section of the silo (GA0 and GA1) recorded throughout the entire
test (maximum for both discharges), for each discharge ow rate.
In contrast to the Type I assays, pressures on the hopper wall
(GA0) increased linearly as the discharge rate increased
(r2 = 0.313), whereas pressures on the cylinder wall (GA1) remained virtually constant, independently of the rate of discharge
(r2 = 0,079).
As in the previous case, it is therefore concluded that a slow discharge does not imply lower pressures than those produced during
a rapid discharge.

Fig. 14 shows the variation in pressures between static states 1

and 2 after a small discharge of material, and the evolution of pressures during the second discharge. It also provides a comparison
between the results obtained here and those estimated using the
calculation method proposed in the Eurocode (CEN, 2006). For
the same reason as in Type I assays, the time intervals at which
pressures were analysed differed for the different outlet diameters
tested. As can be seen in the gure, after a small discharge of material (23%), a variation in pressures occurred inside the silo, and
that when discharge was stopped, pressures did not return to the
initial static state, i.e., the level of pressure reached was maintained in static state 2. As in the previous case, the variation in
pressures after the rst discharge was not progressive along the
height but was more marked and always positive in the lower section of the cylinder at the level of the silo-hopper transition,
whereas in the upper section, the peaks in pressure observed in
the Type I assays appeared. As regards the start of the second discharge, it can be seen that this variation in pressures became even
more pronounced, although the increase or decrease was lower
than that attained in the rst discharge. It can also be seen that this
variation followed the same pattern, i.e., the peaks in pressure
tended to become more marked. In other words, where the pressure had increased before, it now tended to increase even more,
and where before it had decreased, it now tended to decrease even
more. While this was the case for the cylinder wall, it can be seen
in Fig. 14 and from the data shown in Table 2, that at the start of
the second discharge, pressures in the hopper at the silo-hopper
transition (GA0) no longer showed an increase but rather tended
to decrease steadily as the silo was discharged.
Fig. 15 shows the values obtained for the parameters K, nZSK and
l, compared with the values proposed in the Eurocode (CEN,
2006). For the same reason as that discussed for Type I assays,
the values of K and l throughout the test are not shown. In the gure, it can be seen that in static state 2, the value of nZSK was not
constant but rather decreased, due to the phenomenon discussed
above of grain resettling. This parameter attained maximum values
at the beginning of the discharge, for both the discharges that comprised the test: however, at no time were the values proposed in
the Eurocode for this parameter achieved. Focusing on the value
of K, it can be seen that this underwent a sharp increase in the rst
discharge, exceeding the value estimated using the calculation
method proposed in the Eurocode. Furthermore, once discharge
had been halted, this value did not fall to that obtained during

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

53

Fig. 14. Evolution of pressures in the rst and second discharge for the outlet openings tested. Comparison with Eurocode 1, part 4. Type II assays.

Fig. 15. Mean value of resultant vertical stress per unit perimeter at the transition (nzSk,t), lateral pressure ratio at the transition (K), mean value of the wall friction coefcient
between the stored material and the cylinder wall (lm). Comparison between the values obtained in Test 12 and those obtained using Eurocode 1, part 4. Type II assays.

54

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Fig. 16. Normal silo wall pressures (phGA,t) and vertical stress in the stored material at the transition (pvt,t), at each time t. Type III assay, test 19.

the previous static state, which is logical since the normal cylinder
wall pressure at the silo-hopper transition was not recovered
either, whereas vertical stress in the stored material (pvt) did decrease (Fig. 12). As regards the grain to wall coefcient of friction
value, it can be observed that this remained virtually constant
throughout the test, presenting slight temporary increases at the
beginning of each of the discharges.
To summarise, it can be concluded from the results obtained
from the Type II assays that when a small amount of material
was discharged from the silo (in the tests, the amount discharged
ranged from 2% to 3%), a state of overpressure was produced in
the silo with maximum values similar to those attained when a
complete discharge was performed. These pressures barely decreased until the silo was once again empty. Furthermore, even
higher maximum pressures were attained in some zones along
the height of the silo during the second discharge. This aspect
should be taken into account in silo design because the variation
in wall pressures during discharge is not a temporary phenomenon
which disappears when unloading stops, but is a stress state which
persists over time. This is important because in practice, it is common for the material stored to be discharged in stages according to
need or demand, rather than discharging all of it at once.
Fig. 16 shows the curves obtained for the various parameters
discussed above for the Type III assay, and Table 3 shows the
duration of the different stages and their corresponding numerical results. As expected in view of the previous results, following
the rst discharge, static state pressures did not return to the initial values obtained at the end of the rst loading. It can be seen
that pressures at the silo-hopper transition (GA0GA1) decreased
after successive discharges, but this drop was due to progressive
emptying of the silo. Peaks in pressures appeared during the successive discharges performed, which rapidly disappeared again on
returning to static state. As indicated earlier, the peaks observed
during the second discharge in Type II assays were due to
dilatancy.

In other words, a process occurs inside the silo during discharge

which is similar to what would happen if loose granular material
were subjected to shear stress in a triaxial apparatus: the shearing
induces an increase in the materials specic weight, and after successive shearing episodes a weight which could be called critical
is obtained. Once this has happened, further episodes will lead to
variations in volume.
In short, from our interpretation of these results we conclude
that movement of material inside the silo during discharge produces variations in the materials specic weight. There are some
areas in which it becomes compacted, i.e., where it increases its
specic weight, and consequently the pressure exerted on the wall
is also increased, and other areas where its specic weight decreases and therefore the pressure exerted is also reduced. These
areas where specic weight and pressure decrease constitute discontinuities or rupture zones due, on the one hand, to the downward movement of the material in the lower section as a result
of discharge, and on the other, to the slower descent or immobility
of material in the upper section due to grain-wall friction. This
phenomenon would be primarily responsible for the variation in
pressures inside the silo during discharge, since if these variations
were due to the phenomenon known as dilatancy, initial pressures
would be recovered when discharge was halted. This phenomenon
is supported by a detailed examination of the data shown in Table 3. Focusing on the lower section of the silo (PhGA0 to PhGA3), it
can be seen that the greatest variations in pressure occurred during
the rst discharge. Thereafter, slight increases occurred with each
successive discharge, which may be attributed to dilatancy since
previous levels were subsequently recovered.
Focusing on the comparison between the maximum normal
wall pressures at the level of the silo-hopper transition obtained
in tests during discharge and those estimated using Eurocode 1,
part 4, and as can be deduced from Tables 13 and Figs. 10 and
14, the values obtained in tests for normal pressures on the cylinder wall at the level of the silo-hopper transition during discharge

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

were always below those proposed in the Eurocode, as were normal pressures on the hopper wall at the same level, although in
this case the values were much more similar.
4. Conclusions
Pressures are not constant in the static state, since the grain
undergoes a process of resettlement which occurs at increasingly
spaced intervals.
In the tests conducted, a reduced ow rate (kg/s) of the granular
material during discharge did not entail an associated decrease in
pressures on the silo wall.
At the beginning of discharge, the granular material underwent
a process of variation in the unit weight, increasing or even
decreasing its unit weight in some areas more than others, and it
was this effect which was primarily responsible for the variations
in pressure inside the silo during discharge. Even if discharge is
halted at this point, the silo will not recover the previous pressure
values, reaching a state of overpressure similar to that produced
when performing a complete discharge in one step.
Once the material has reached a certain unit weight, which
could be called critical, variations in pressure during discharge
are due to the phenomenon known as dilatancy, i.e., the increase
in volume caused by dilatancy in a vertical direction due to the
presence of stiff walls.
During discharge, the normal hopper wall pressures at the level
of the silo-hopper transition are close to those estimated using the
calculation method proposed in Eurocode 1, part 4, although this
latter always produces higher values.
Acknowledgements
The authors thank the Spanish Research and Technology Commission (CICYT) (Research Project AGL2005-07430-C02-01/AGR)
and the Regional Executive of Castile and Len (Research Project
LE020A10-2) for nancing this research.
References
Ahn, H., Basaranoglu, Z., Yilmaz, M., Bugutekin, A., Gl, M.Z., 2008. Experimental
investigation of granular ow through an orice. Powder Technology 186, 65
71.
Artoni, R., Santomaso, A., Canu, P., 2009. Simulation of dense granular ows:
dynamics of wall stress in silos. Chemical Engineering Science 64, 40404050.
Askegaard, V., Munch-Andersen, J., 1985. Results from tests with normal and shear
stress cells in a medium-scale model silo. Powder Technology 44, 151157.
Ayuga, F., 2008. Some Unresolved Problems in the Design of Steel Cylindrical Silos.
Crc Press-Taylor & Francis Group, Boca Raton.
Ayuga, F., Guaita, M., Aguado, P., 2001a. Static and dynamic silo loads using nite
element models. Journal of Agricultural Engineering Research, 78.
Ayuga, F., Guaita, M., Aguado, P.J., Couto, A., 2001b. Discharge and the eccentricity of
the hopper inuence on the silo wall pressures. Journal of Engineering
Mechanics Asce 127, 10671074.
Balevicius, R., Sielamowicz, I., Mrz, Z., Kacianauskas, R., 2011. Investigation of wall
stress and outow rate in a at-bottomed bin: a comparison of the DEM model
results with the experimental measurements. Powder Technology 214, 322
336.
Briassoulis, D., 2000. Finite element analysis of a cylindrical silo shell under
unsymmetrical pressure distributions. Computers & Structures 78, 271281.
Brown, C.J., Lahlouh, E.H., Rotter, J.M., 2000. Experiments on a square planform steel
silo. Chemical Engineering Science 55, 43994413.
CEN, 2006. EN 1991-4:2006. Eurocode 1: Actions on Structures. Part 4: Silos and
Tanks. European Committee for Standardization, Brussels.
Chen, J.F., Rotter, J.M., Ooi, J.Y., 1998. Statistical inference of unsymmetrical silo
pressures from comprehensive wall strain measurements. Thin-Walled
Structures 31, 117136.
Chen, J.F., Rotter, J.M., Ooi, J.Y., Zhong, Z., 2005. Flow pattern measurement in a full
scale silo containing iron ore. Chemical Engineering Science 60, 30293041.
Chen, J.F., Rotter, J.M., Ooi, J.Y., Zhong, Z., 2007. Correlation between the ow pattern
and wall pressures in a full scale experimental silo. Engineering Structures 29,
23082320.
Chou, C.S., Hsu, J.Y., Lau, Y.D., 2002. The granular ow in a two-dimensional atbottomed hopper with eccentric discharge. Physica A: Statistical Mechanics and
its Applications 308, 4658.

55

Coetzee, C.J., Els, D.N.J., 2009. Calibration of discrete element parameters and the
modelling of silo discharge and bucket lling. Computers and Electronics in
Agriculture 65, 198212.
Couto, A., Ruiz, A., Aguado, P.J., 2012. Design and instrumentation of a mid-size test
station for measuring static and dynamic pressures in silos under different
conditions Part I: description. Computers and Electronics in Agriculture 85,
164173.
Ding, S., Rotter, J.M., Ooi, J.Y., Enstad, G., 2011. Development of normal pressure and
frictional traction along the walls of a steep conical hopper during lling. ThinWalled Structures 49, 12461250.
Dogangun, A., Karaca, Z., Durmus, A., Sezen, H., 2009. Cause of damage and failures
in silo structures. Journal of Performance of Constructed Facilities 23, 6571.
Gillie, M., Holst, J., 2003. Structural behaviour of silos supported on discrete,
eccentric brackets. Journal of Constructional Steel Research 59, 887910.
Goodey, R.J., Brown, C.J., 2004. The inuence of the base boundary condition in
modelling lling of a metal silo. Computers & Structures 82, 567579.
Grudzien, K., Niedostatkiewicz, M., Adrien, J., Tejchman, J., Maire, E., 2011.
Quantitative estimation of volume changes of granular materials during silo
ow using X-ray tomography. Chemical Engineering and Processing: Process
Intensication 50, 5967.
Guaita, M., Couto, A., Ayuga, F., 2003. Numerical simulation of wall pressure during
discharge of granular material from cylindrical silos with eccentric hoppers.
Biosystems Engineering 85, 101109.
Hrtl, J., Ooi, J.Y., Rotter, J.M., Wojcik, M., Ding, S., Enstad, G.G., 2008. The inuence
of a cone-in-cone insert on ow pattern and wall pressure in a full-scale silo.
Chemical Engineering Research and Design 86, 370378.
Janssen, H.A., 1895. Versuch ber Getreidedruck in Sillozellen. Zeitschrift des
Vereins Deutscher Ingenieure 39, 10451049.
Jenike, A.W., Johanson, J.R., 1969. On theory of bin loads. Mechanical Engineering
91, 43&.
Juan, A., Moran, J.M., Guerra, M.I., Couto, A., Ayuga, F., Aguado, P.J., 2006.
Establishing stress state of cylindrical metal silos using nite element
method: comparison with ENV 1993. Thin-Walled Structures 44, 11921200.
Khelil, A., 2002. Buckling of steel shells subjected to non-uniform axial and pressure
loading. Thin-Walled Structures 40, 955970.
Martinez, M.A., Alfaro, I., Doblare, M., 2002. Simulation of axisymmetric discharging
in metallic silos. Analysis of the induced pressure distribution and comparison
with different standards. Engineering Structures 24, 15611574.
Matchett, A.J., ONeill, J.C., Shaw, A.P., 2008. Stress distributions in 2-dimensional,
wedge hoppers with circular arc stress orientation a co-ordinate-specic
LamMaxwell model. Powder Technology 187, 298306.
Moya, M., Ayuga, F., Guaita, M., Aguado, P., 2002. Mechanical properties of granular
agricultural materials. Transactions of the Asae 45, 15691577.
Nielsen, J., 2008. From silo phenomena to load models. In: Chen, J.F., Teng, J.G. (Eds.),
International Conference on Structures and Granular Solids - From Scientic
Principles to Engineering Applications, Structures and Granular Solids: From
Scientic Principles to Engineering Applications, JUL 01-02, 2008, Royal Soc
Edinburgh,
Edinburgh,
Scotland,
4957.
http://dx.doi.org/10.1201/
9780203884447.ch4.
Nielsen, J., Askegaard, V., 1977. Scale errors in model tests on granular media with
special reference to silo models. Powder Technology 16, 123130.
Ooi, J.Y., Pham, L., Rotter, J.M., 1990. Systematic and random features of measured
pressures on full-scale silo walls. Engineering Structures 12, 7487.
Pieper, K., Wenzel, F., 1964. Druckverhaltnisse in Silozellen. Verlag von Wilhelm
Ernst & Sohn, Berlin, Munich.
Ramirez, A., Moya, M., Ayuga, F., 2009. Determination of the mechanical properties
of powdered agricultural products and sugar. Particle & Particle Systems
Characterization 26, 220230.
Ramirez, A., Nielsen, J., Ayuga, F., 2010. Pressure measurements in steel silos with
eccentric hoppers. Powder Technology 201, 720.
Rotter, J.M., Brown, C.J., Lahlouh, E.H., 2002. Patterns of wall pressure on lling a
square planform steel silo. Engineering Structures 24, 135150.
Rotter, J.M., Jumikis, P.T., Fleming, S.P., Porter, S.J., 1989. Experiments on the
buckling of thin-walled model silo structures. Journal of Constructional Steel
Research 13, 271299.
Ruiz, A., Couto, A., Aguado, P., 2012a. Design and instrumentation of a mid-size test
station for measuring static and dynamic pressures in silos under different
conditions Part II: construction and validation. Computers and Electronics in
Agriculture.
Ruiz, A., Couto, A., Aguado, P.J., 2012b. Design and instrumentation of a mid-size
test station for measuring static and dynamic pressures in silos under different
conditions Part II: construction and validation. Computers and Electronics in
Agriculture 85, 174187.
Sadowski, A.J., Rotter, J.M., 2011. Steel silos with different aspect ratios: I
behaviour under concentric discharge. Journal of Constructional Steel Research
67, 15371544.
Schurich, T., Frll, C., Enstad, G.G., 2001. Full scale silo tests and numerical
simulations of the cone in cone concept for mass ow. In: Kalman, A.L.a.H.
(Ed.), Handbook of Powder Technology. Elsevier Science B.V., pp. 175180.
Sielamowicz, I., Blonski, S., Kowalewski, T.A., 2005. Optical technique DPIV in
measurements of granular material ows, Part 1 of 3 plane hoppers. Chemical
Engineering Science 60, 589598.
Sielamowicz, I., Czech, M., Kowalewski, T.A., 2011. Empirical description of granular
ow inside a model silo with vertical walls. Biosystems Engineering 108, 334
344.

56

A. Couto et al. / Computers and Electronics in Agriculture 96 (2013) 4056

Slominski, C., Niedostatkiewicz, M., Tejchman, J., 2007. Application of particle image
velocimetry (PIV) for deformation measurement during granular silo ow.
Powder Technology 173, 118.
Song, C.Y., 2004. Effects of patch loads on structural behavior of circular atbottomed steel silos. Thin-Walled Structures 42, 15191542.
Sugita, M., 1972. Flow and pressures of noncohesive granular materials in funnelow bins. Mechanical Engineering 94, 60&.
Tejchman, J., Ummenhofer, T., 2000. Bedding effects in bulk solids in silos:
experiments and a polar hypoplastic approach. Thin-Walled Structures 37,
333361.
Teng, J.G., Chan, F., 1999. New buckling approximations for T-section ringbeams
simply-supported at inner edge. Engineering Structures 21, 889897.
Teng, J.G., Chan, F., 2001. Plastic buckling strength of T-section transition ringbeams
in steel silos and tanks. Engineering Structures 23, 280297.
Teng, J.G., Lin, X., Rotter, J.M., Ding, X.L., 2005. Analysis of geometric imperfections
in full-scale welded steel silos. Engineering Structures 27, 938950.
Vidal, P., Couto, A., Ayuga, F., Guaita, M., 2006. Inuence of hopper eccentricity on
discharge of cylindrical mass ow silos with rigid walls. Journal of Engineering
Mechanics Asce 132, 10261033.

Vidal, P., Gallego, E., Guaita, M., Ayuga, F., 2008. Finite element analysis under
different boundary conditions of the lling of cylindrical steel silos having an
eccentric hopper. Journal of Constructional Steel Research 64, 480492.
Wu, J., Binbo, J., Chen, J., Yang, Y., 2009. Multi-scale study of particle ow in silos.
Advanced Powder Technology 20, 6273.
Wjcik, M., Tejchman, J., 2009. Modeling of shear localization during conned
granular ow in silos within non-local hypoplasticity. Powder Technology 192,
298310.
Wjcik, M., Tejchman, J., Enstad, G.G., 2012. Conned granular ow in silos with
inserts. Full-scale experiments. Powder Technology 222, 1536.
Yang, S.-C., Hsiau, S.-S., 2001. The simulation and experimental study of granular
materials discharged from a silo with the placement of inserts. Powder
Technology 120, 244255.
Yang, Y., Rotter, M., Ooi, J., Wang, Y., 2011. Flow channel boundaries in silos.
Chemical Engineering & Technology 34, 12951302.
Zhong, Z., Ooi, J.Y., Rotter, J.M., 2001. The sensitivity of silo ow and wall stresses to
lling method. Engineering Structures 23, 756767.